Change the denominator, change the score.
Planned downtime, breaks, maintenance windows: decide what counts as available time and OEE moves by double digits without a single extra part leaving the plant.
OEE · AVAILABILITY × PERFORMANCE × QUALITY
OEE software that tells you what to do with the number
A plant can improve its OEE score and produce exactly the same amount. It happens more often than anyone admits.
THE SHORT ANSWER
OEE is availability × performance × quality, and it is a scoreboard, not a strategy. The number is only useful when the three losses are separated on trustworthy data, compared machine against its own history, and fed back into scheduling — otherwise it measures how the plant reports, not how the plant runs.
HOW OEE GETS GAMED WITHOUT ANYONE LYING
One number hides three different problems.
An 65% OEE from availability losses and an 65% OEE from quality losses need opposite actions. The headline figure treats them as the same illness.
Comparing machines that were never comparable.
Ranking a 20-year-old press against a new line produces a league table, not an insight. The useful comparison is each machine against its own best behaviour.
The market reports OEE.
Atherya reads the losses behind it and puts them back into the plan.
WHAT ATHERYA DOES INSTEAD
Calculated from the data you already write.
Cycle records, state changes and quality results from historian, MES and ERP. No parallel data-entry ritual to keep the score alive.
Each machine against itself.
The reference is that machine's learned normal, per recipe and per shift — the only comparison where a gap means something you can act on.
Losses ranked by recoverable hours.
Not the worst percentage: the largest number of hours you can realistically get back, given the bottleneck and the order book.
The gap becomes a scheduling proposal.
Atherya proposes the maintenance window or the sequence change that closes the loss, against the live plan, and waits for approval.
READ FROM ONE REAL PLANT — BEFORE INSTALL
6years of raw history6Mreadings · 42 machines · 5 families141machine × recipe signatures98 → 7open questions closed from data
Rubber moulding, northern Italy. Figures from a real onboarding, read from existing data before any installation. Anything beyond them is a scenario, and we declare it as a scenario.
AN OEE READ ON A REAL PLANT
- Six years of history, roughly 6 million readings, 42 machines across 5 asset families — read before installation.
- The plant's real average sat at 47%, with the best machine far above it: the spread between machines was a bigger opportunity than the average itself.
- 60.8% of the bottleneck concentrated on the presses, and the same recipe ran 5.9% better at night — both invisible in a single plant-level OEE figure.
What makes Atherya different
Most predictive maintenance tools score signals. Atherya builds an operational memory of the plant first, then reasons on top of it — with the evidence in plain sight and the decision left to a person.
Operational memory
Every event, anomaly, intervention and shift becomes shared memory. What was a log yesterday is experience tomorrow, and the model reads new signals against it.
Déjà Vu: it has seen this before
Not just "anomaly detected", but when it already happened, how similar it was, how it evolved and which action worked. The senior maintainer's memory, available to everyone.
Context that kills false alarms
Machine defects, weak points, how the crew actually works and the environment around the line. A deviation that is normal for that machine, that shift or that season stays quiet.
Living FMEA
FMEA, manuals and procedures become an active part of the reasoning: causes, effects, sensors and suggested actions are connected, instead of sitting in a document nobody opens.
Explainable, not magic
Every alert arrives with the signals involved, the comparable history, the failure mode and a confidence level. Data, interpretation and decision stay separate and verifiable.
It acts, with your sign-off
Atherya does not stop at the warning: it proposes the maintenance window against real orders and shifts, and replans when something changes. Nothing is applied until a person approves it.
On the data you already have
It works on PLC, SCADA, MES, ERP, maintenance records and feedback as they are — fragmented and legacy included. Sensors are added only where no existing signal carries the degradation.
One brain, not a maintenance silo
Maintenance, production and planning read the same operational state, so a predicted failure immediately becomes a scheduling question instead of a separate dashboard.
QUESTIONS, ANSWERED
Straight answers.
How is OEE calculated?
Availability × performance × quality. Availability is run time over planned production time, performance is actual output over theoretical output at ideal cycle, quality is good parts over total parts. The definitions of planned time and ideal cycle are where most disagreements start.
What is a good OEE?
It depends on the process and the asset mix, so a single benchmark number is usually marketing. The honest comparison is your own trend and the spread between your machines, not another plant's headline figure.
Can OEE be calculated automatically?
Yes, when the state changes, cycle counts and quality results already exist in the historian, MES or ERP. Atherya reads those sources directly, so the score does not depend on manual entry.
Why does a higher OEE not always mean more output?
Because the score is sensitive to how planned time is defined and to which machines are included. Output changes when the constraint moves — which is why the losses matter more than the headline.
Does Atherya replace my MES?
No. It sits on top of the existing stack — MES, ERP, historian, CMMS — and adds the model of the plant and the decision layer above them.
YOUR PLANT · YOUR DATA · THE PROOF
Get the losses behind your number.
Send the export. We separate the three losses and show which hours are recoverable.
Discover Atherya